Word embedding

About: Word embedding is a research topic. Over the lifetime, 4683 publications have been published within this topic receiving 153378 citations. The topic is also known as: word embeddings.

Intrinsic and extrinsic evaluations of word embeddings

Abstract: In this paper, we first analyze the semantic composition of word embeddings by cross-referencing their clusters with the manual lexical database, WordNet. We then evaluate a variety of word embedding approaches by comparing their contributions to two NLP tasks. Our experiments show that the word embedding clusters give high correlations to the synonym and hyponym sets in WordNet, and give 0.88% and 0.17% absolute improvements in accuracy to named entity recognition and part-of-speech tagging, respectively.

Toward Word Embedding for Personalized Information Retrieval

Abstract: This paper presents preliminary works on using Word Embedding (word2vec) for query expansion in the context of Personalized Information Retrieval. Traditionally, word embeddings are learned on a general corpus, like Wikipedia. In this work we try to personalize the word embeddings learning, by achieving the learning on the user's profile. The word embeddings are then in the same context than the user interests. Our proposal is evaluated on the CLEF Social Book Search 2016 collection. The results obtained show that some efforts should be made in the way to apply Word Embedding in the context of Personalized Information Retrieval.

A Superior Arabic Text Categorization Deep Model (SATCDM)

Abstract: Categorizing Arabic text documents is considered an important research topic in the field of Natural Language Processing (NLP) and Machine Learning (ML). The number of Arabic documents is tremendously increasing daily as new web pages, news articles, social media contents are added. Hence, classifying such documents in specific classes is of high importance to many people and applications. Convolutional Neural Network (CNN) is a class of deep learning that has been shown to be useful for many NLP tasks, including text translation and text categorization for the English language. Word embedding is a text representation currently used to represent text terms as real-valued vectors in vector space that represent both syntactic and semantic traits of text. Current research studies in classifying Arabic text documents use traditional text representation such as bag-of-words and TF-IDF weighting, but few use word embedding. Traditional ML algorithms have already been used in Arabic text categorization, and good results are achieved. In this study, we present a Multi-Kernel CNN model for classifying Arabic news documents enriched with n-gram word embedding, which we call A Superior Arabic Text Categorization Deep Model (SATCDM). The proposed solution achieves very high accuracy compared to current research in Arabic text categorization using 15 of freely available datasets. The model achieves an accuracy ranging from 97.58% to 99.90%, which is superior to similar studies on the Arabic document classification task.

Constraining Word Embeddings by Prior Knowledge – Application to Medical Information Retrieval

Abstract: Word embedding has been used in many NLP tasks and showed some capability to capture semantic features. It has also been used in several recent studies in IR. However, word embeddings trained in unsupervised manner may fail to capture some of the semantic relations in a specific area (e.g. healthcare). In this paper, we leverage the existing knowledge (word relations) in the medical domain to constrain word embeddings using the principle that related words should have similar embeddings. The resulting constrained word embeddings are used to rerank documents, showing superior effectiveness to unsupervised word embeddings.